Pär Jonsén

Identification of lumped parameter automotive crash models for bumper system development

During the design and development process of bumper systems for the automotive industry, information about the future car model is limited. Normally, iterative finite element (FE) analyses of different crash loading tests are used to find an appropriate bumper system to the coming car model. Because of the lack of information, only a rough model of the car is normally utilised in the FE simulations. This leads to uncertainties in the bumper design since the dynamic response of the car is dependent on the load case and the properties of the actual bumper system. This paper presents a method for identification of lumped parameter models based on results from crash tests of a Volvo S40. The ability to predict the measured results for models with different number of degrees of freedom (DOF) is investigated. Also, a validation of the model together with an FE mesh of the bumper system is presented. The results clearly show that a linear mass spring damper model with 2 DOF can be used to predict the response from the measurements in case of symmetric loading. Further increase of the number of DOF only causes small or no improvements of the agreement between the predicted and measured crash response.

Green Body Behaviour of High Velocity Pressed Metal Powder

High velocity compaction (HVC) is a production technique with capacity to significantly improve the mechanical properties of powder metallurgy (PM) parts. Several investigations indicate that high-density components can by obtained using HVC. Other characteristics are low ejection force and uniform density. Investigated here are green body data such as density, tensile strength, radial springback, ejection force and surface flatness. Comparisons are performed with conventional compaction using the same pressing conditions. Cylindrical samples of a pre-alloyed water atomized iron powder are used in this experimental investigation. The different behaviour of HVC-pressed green bodies compared to conventional pressed green bodies are analysed and discussed. The HVC process in this study resulted in a better compressibility curve and lower ejection force compared to conventional quasi static pressing. Vertical scanning interferometry (VSI) measurements show that the HVC process gives flatter sample surfaces.

Modelling and simulation of explosions in soil interacting with deformable structures

A detonating explosive interacting with a deformable structure is a highly transient and non-linear event. In field blast trials of military vehicles, a standard procedure is often followed in order to reduce the uncertainties and increase the quality of the test. If the explosive is buried in the ground, the state of the soil must meet specific demands. In the present work, laboratory experiments have been performed to characterize the behaviour of a soil material. Soil may be considered a three-phase medium, consisting of solid grains, water and air. Variations between the amounts of these phases affect the mechanical properties of the soil. The experimental outcome has formed input data to represent the soil behaviour included in a three-phase elastic-plastic cap model. This unified constitutive model for soil has been used for numerical simulations representing field blast trials, where the explosive load is interacting with a deformable structure. The blast trials included explosive buried at different depths in wet or dry sand. A dependence of the soil initial conditions can be shown, both in the past field trials along with the numerical simulations. Even though some deviations exist, the simulations showed in general acceptable agreement with the experimental results.

Correlation of vehicle crash model parameters to car properties in low-speed collisions: a design of experiments approach

In the current study, a methodology for relating model parameters in a one dimensional Mass Spring Damper (MSD) model to global properties of a car, e.g. axial stiffness, bending stiffness and mass, is presented. It is shown that these three vehicle properties affect the vehicles crash performance in low-speed collision tests used for industrial verification of bumper system performance. Based on information of the properties for a vehicle under development, parameters in the MSD model can be adjusted to give the correct boundary conditions for a finite element (FE) crash simulation with a candidate bumper design. In the FE simulations, the MSD model is then coupled to the FE mesh of candidate bumper design to find a bumper that meets the crash performance requirements of a car under development. The methodology is based on Design of Experiments (DOE) and FE simulations on a public domain model of a Ford Taurus. The knowledge gained from this study gives a valuable tool to use in design and development of bumper systems for the automotive industry.

The Effects of Surface Topography and Lack of Fusion on The Fatigue Strength of Laser Hybrid Welds

The geometrical aspects of laser hybrid welds (before, during and after the process) differ from autonomous laser welding and from arc welding. When studying the fatigue behaviour of laser hybrid w ...

Mechanical characterization and modelling of the temperature-dependent impact behaviour of a biocompatible poly(L-lactide)/poly(ε-caprolactone) polymer blend.

Poly(ε-caprolactone) (PCL) is a ductile, bioabsorbable polymer that has been employed as a blend partner for poly(L-lactic acid) (PLLA). An improvement of the material strength and impact resistance of PLLA/PCL polymer blends compared to pure PLLA has been shown previously. To use numerical simulations in the design process of new components composed of the PLLA/PCL blend, a constitutive model for the material has to be established. In this work, a constitutive model for a PLLA/PCL polymer blend is established from the results of compressive tests at high and low strain rates at three different temperatures, including the body temperature. Finite element simulations of the split Hopkinson pressure bar test using the established constitutive model are carried out under the same condition as the experiments. During the experiments, the changes in the diameter and thickness of the specimens are captured by a high-speed video camera. The accuracy of the numerical model is tested by comparing the simulation results, such as the stress, strain, thickness and diameter histories of the specimens, with those measured in the experiments. The numerical model is also validated against an impact test of non-homogenous strains and strain rates. The results of this study provide a validated numerical model for a PLLA/PCL polymer blend at strain rates of up to 1800 s(-1) in the temperature range between 22°C and 50°C.

Modelling and Simulation of Tensile Fracture in High Velocity Compacted Metal Powder

In cold uniaxial powder compaction, powder is formed into a desired shape with rigid tools and a die. After pressing, but before sintering, the compacted powder is called green body. A critical property in the metal powder pressing process is the mechanical properties of the green body. Beyond a green body free from defects, desired properties are high strength and uniform density. High velocity compaction (HVC) using a hydraulic operated hammer is a production method to form powder utilizing a shock wave. Pre‐alloyed water atomised iron powder has been HVC‐formed into circular discs with high densities. The diametral compression test also called the Brazilian disc test is an established method to measure tensile strength in low strength material like e.g. rock, concrete, polymers and ceramics. During the test a thin disc is compressed across the diameter to failure. The compression induces a tensile stress perpendicular to the compressed diameter. In this study the test have been used to study crack init...

Generation of segmental chips in metal cutting modeled with the PFEM

The Particle Finite Element Method, a lagrangian finite element method based on a continuous Delaunay re-triangulation of the domain, is used to study machining of Ti6Al4V. In this work the method is revised and applied to study the influence of the cutting speed on the cutting force and the chip formation process. A parametric methodology for the detection and treatment of the rigid tool contact is presented. The adaptive insertion and removal of particles are developed and employed in order to sidestep the difficulties associated with mesh distortion, shear localization as well as for resolving the fine-scale features of the solution. The performance of PFEM is studied with a set of different two-dimensional orthogonal cutting tests. It is shown that, despite its Lagrangian nature, the proposed combined finite element-particle method is well suited for large deformation metal cutting problems with continuous chip and serrated chip formation.

A Particle Finite Element Method for Machining Simulations

Metal cutting process is a nonlinear dynamic problem that includes geometrical, material, and contact nonlinearities. In this work a Lagrangian finite element approach for simulation of metal cutti ...

High pressure characterization and modelling of CaCO3 powder mix in the Bridgman anvil apparatus

For investigating high pressure sintering processes, numerical models can be used. This will demand material models which give realistic mechanical response throughout the whole parameter space of the actual process. As the pressures become higher, the material density approaches its full theoretical value and the elastic part of the material properties becomes increasingly important. In this investigation, Poissons ratio was determined using ultrasonic pulse-echo measurements. A new elastic model and an improved plasticity model were implemented into a user-defined material subroutine in a finite element (FE) code. To experimentally investigate the load displacement response and pressure distribution in powder compacts during pressing, a pressure instrumented Bridgman anvil apparatus was used. Validation of the FE model was conducted against experimental data from pressing experiments using two different start densities. The results show that the simulation model is indeed capable of reproducing load–thickness curves and pressure profiles reasonable close to the experimental curves.